Fast Low-light Image Enhancement Algorithm Based on Fusion

Abstract

Low-light images generally exist in various situations, and the existing low-light image enhancement algorithms are poor in generality and have high computational complexity, and the comfort of the enhanced image is not satisfactory. This paper proposes a fast low-light image enhancement algorithm based on fusion. First, transform the passing space to HSV space to extract the V component. Then, according to the influence of Weber-Fechner's law on human visual senses, and effectively combining the characteristics of low-light images, a global adaptive brightness transformation function is constructed, and this function generates the first enhanced image. Then use CLAHE to increase the local contrast and generate a second enhanced image. Finally, an image fusion strategy is adopted to extract image details to fuse the image effectively. The proposed algorithm is fast and straightforward, can effectively improve the quality of low-light images, and better preserve its naturalness. Experimental results show that the proposed algorithm has achieved good results both subjectively and objectively.